Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Message-ID: <924be4021c3b9f65d15aa12765facb9eb58911b0.camel@redhat.com>
Subject: Re: [RFC] trace: Add option for polling ring buffers
From:   Nicolas Saenz Julienne <nsaenzju@redhat.com>
To:     Steven Rostedt <rostedt@goodmis.org>
Cc:     linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
        mingo@redhat.com, corbet@lwn.net, mtosatti@redhat.com
Date:   Wed, 02 Jun 2021 11:38:03 +0200
In-Reply-To: <20210528133253.27c749ab@gandalf.local.home>
References: <20210519175755.670876-1-nsaenzju@redhat.com>
         <20210528133253.27c749ab@gandalf.local.home>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.40.0 (3.40.0-1.fc34) 
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi Steven, thanks for having a look at this.

On Fri, 2021-05-28 at 13:32 -0400, Steven Rostedt wrote:
> On Wed, 19 May 2021 19:57:55 +0200
> Nicolas Saenz Julienne <nsaenzju@redhat.com> wrote:
> 
> > To minimize trace's effect on isolated CPUs. That is, CPUs were only a
> > handful or a single, process are allowed to run. Introduce a new trace
> > option: 'poll-rb'.
> > 
> > This option changes the heuristic used to wait for data on trace
> > buffers. The default one, based on wait queues, will trigger an IPI[1]
> > on the CPU responsible for new data, which will take care of waking up
> > the trace gathering process (generally trace-cmd). Whereas with
> > 'poll-rb' we will poll (as in busy-wait) the ring buffers from the trace
> > gathering process, releasing the CPUs writing trace data from doing any
> > wakeup work.
> > 
> > This wakeup work, although negligible in the vast majority of workloads,
> > may cause unwarranted latencies on systems running trace on isolated
> > CPUs. This is made worse on PREEMPT_RT kernels, as they defer the IPI
> > handling into a kernel thread, forcing unwarranted context switches on
> > otherwise extremely busy CPUs.
> > 
> > To illustrate this, tracing with PREEMPT_RT=y on an isolated CPU with a
> > single process pinned to it (NO_HZ_FULL=y, and plenty more isolation
> > options enabled). I see:
> >   - 50-100us latency spikes with the default trace-cmd options
> >   - 14-10us latency spikes with 'poll-rb'
> >   - 11-8us latency spikes with no tracing at all
> > 
> > The obvious drawback of 'poll-rb' is putting more pressure on the
> > housekeeping CPUs. Wasting cycles. Hence the notice in the documentation
> > discouraging its use in general.
> > 
> > [1] The IPI, in this case, an irq_work, is needed since trace might run
> > in NMI context. Which is not suitable for wake-ups.
> 
> Can't this simply be done in user-space?
>
> Set the reading of the trace buffers to O_NONBLOCK and it wont wait for
> buffering to happen, and should prevent it from causing the IPI wake ups.

Yes, I hadn't thought of O_NONBLOCK...

> If you need this for trace-cmd, we can add a --poll option that would do
> this.

I just sent a patch with my attempt at implementing --poll.

-- 
Nicolás Sáenz